ODP: thru zero VCO questions

[Martin Czech]

:::Hi Martin:
:::
:::You can divide down the quartz oscillators to around 50 KHz, the filtering
:::will be easier there.  If a master clock is used to divide from, and it
:::supplies both mixing frequencies, frequency errors will largely cancel out
:::after translating the VCO up and then down.   Or, an image rejection mixer
:::using 90 degree hybrids can be used, as you indicated.   There are many
:::possible ways to do this.
:::
:::Terry Michaels

Hi Terry:

I think we want to do a frequency shifter in order to use it as fm
device (?). This would require the mixing frequencys to be differnt by
some ammount, this could be done via quartz and plls but is certainly
not easy. And how would one perform fast fm?

And the sideband rejection question remains, as you said. If you allow
40Hz for the lowest signal, the sidebands are only 80 Hz away from each
other. Even at 50kHz this is a major problem, 80Hz/50KHz, wouldnt that
require a Q of > 600 as an initial guess for an LC circuit (that obviously
would't do what we want it to do).

I'd buy all that (with phase method) if I where to fm live sound and
such, but not for fm simple waves. I feel that this will make very much
more trouble then the original problem with the up-down ff uncertaincy
we started with.

Or is there anything is missed?

m.c.